© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 81624
    Prev
    Next
    INDEX
    Explanations

    populist pursuit

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron’s activations rise in proportion to the length of the token, so it’s basically detecting longer words.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Slut
    -0.96
    listview
    -0.91
    Brisbane
    -0.88
    xox
    -0.83
    ären
    -0.81
     endast
    -0.80
     lekker
    -0.79
    のでは
    -0.79
    officially
    -0.79
     vụ
    -0.79
    POSITIVE LOGITS
    collegiate
    1.02
    Column
    0.97
     Column
    0.93
     Kedua
    0.92
     Colleagues
    0.91
     diverso
    0.91
     zaby
    0.88
     extraordinarily
    0.87
     ظرف
    0.87
    鈥
    0.87
    Activations Density 0.013%

    No Known Activations